Modeling and forecasting the oil volatility index

• Author: João H. Gonçalves Mazzeu,Helena Veiga,Massimo B. Mariti
• Date: 01 December 2019
• Published date: 01 December 2019
• DOI: http://doi.org/10.1002/for.2598

Received: 15 April 2018 Revised: 15 January 2019 Accepted: 11 April 2019

DOI: 10.1002/for.2598

RESEARCH ARTICLE

Modeling and forecasting the oil volatility index

João H. Gonçalves Mazzeu1Helena Veiga2,3 Massimo B. Mariti4

1Department of Statistics, University of

Campinas, Campinas-SP,Brazil

2Department of Statistics and Instituto

Flores de Lemus, Universidad Carlos III

de Madrid, Spain

3BRU-IUL, Instituto Universitário de

Lisboa, Portugal

4University Milano-Bicocca, Italy

Correspondence

Helena Veiga, Department of Statistics

and Instituto Flores de Lemus,

Universidad Carlos III de Madrid, Spain.

Email: mhveiga@est-econ.uc3m.es

Funding information

Coordenação de Aperfeiçoamento de

Pessoal de Nível Superior, Grant/Award

Number: 88882.305837/2018-01;

Fundação para a Ciência e a Tecnologia,

Grant/AwardNumber:

UID/GES/00315/2019; Ministerio de

Economía y Competitividad, Grant/Award

Number: PGC2018-096977-B-I00 and

ECO2015-70331-C2-2-R

Abstract

The increase in oil price volatility in recent years has raised the importance of

forecasting it accurately for valuing and hedging investments.The paper models

and forecasts the crude oil exchange-traded funds (ETF) volatility index, which

has been used in the last years as an important alternative measure to track

and analyze the volatility of future oil prices. Analysis of the oil volatility index

suggests that it presents features similar to those of the daily market volatility

index, such as long memory,which is modeled using well-known heterogeneous

autoregressive (HAR) specifications and new extensions that are based on net

and scaled measures of oil price changes. The aim is to improve the forecasting

performance of the traditional HAR models by including predictors that capture

the impact of oil price changes on the economy. The performance of the new

proposals and benchmarks is evaluated with the model confidence set (MCS)

and the Generalized-AutoContouR(G-ACR) tests in terms of point forecasts and

density forecasting, respectively.We find that including the leverage in the con-

ditional mean or variance of the basic HAR model increases its predictive ability.

Furthermore, when considering density forecasting, the best models are a con-

ditional heteroskedastic HAR model that includes a scaled measure of oil price

changes, and a HAR model with errors following an exponential generalized

autoregressive conditional heteroskedasticity specification. In both cases, we

consider a flexible distribution for the errors of the conditional heteroskedastic

process.

KEYWORDS

forecasting oil volatility, heterogeneous autoregression, leverage, net oil price changes, scaled oil

price changes

1INTRODUCTION

The price of oil has been fluctuating dramatically in the

last decade. It reached its maximum price in July 2008, to

plunge to a value of $30.28 per barrel some months later.In

the last 7 years a barrel of crude oil has ranged from $125 to

$30. These oil price fluctuations increase oil volatility and,

consequently, the risk exposure of companies dedicated to

exploring for and processing oil, and investors.

In the last decade investments in commodities have

grown quickly, with oil accounting for a high percentage

of these investments; see Baffes (2007). Given the impor-

tance of oil among the commodities, the Chicago Board

Option Exchange (CBOE) has calculated and reported the

crude oil ETF volatility index since May 2007. According

to CBOE, the index “measures the market's expectation

of 30-day volatility of crude oil prices and it is calculated

using the market volatility index (VIX) methodology for

Journal of Forecasting. 2019;38:773–787. wileyonlinelibrary.com/journal/for © 2019 John Wiley & Sons, Ltd. 773